Update config.json
#1
by
sedrickkeh
- opened
- config.json +2 -4
config.json
CHANGED
@@ -4,7 +4,6 @@
|
|
4 |
],
|
5 |
"model_type": "openlm",
|
6 |
"params": null,
|
7 |
-
"params_args_dict": {
|
8 |
"apply_qk_norm": true,
|
9 |
"attn_activation": null,
|
10 |
"attn_name": "auto",
|
@@ -13,7 +12,7 @@
|
|
13 |
"dim": 2048,
|
14 |
"ffn_type": "swiglu_torch",
|
15 |
"model": "open_lm_1b_swiglutorch",
|
16 |
-
"
|
17 |
"moe_capacity_factor": 1.25,
|
18 |
"moe_expert_model_parallelism": false,
|
19 |
"moe_freq": 0,
|
@@ -29,8 +28,7 @@
|
|
29 |
"qk_norm": true,
|
30 |
"seq_len": 2048,
|
31 |
"vocab_size": 50432,
|
32 |
-
"weight_tying": false
|
33 |
-
},
|
34 |
"torch_dtype": "float32",
|
35 |
"transformers_version": "4.40.2"
|
36 |
}
|
|
|
4 |
],
|
5 |
"model_type": "openlm",
|
6 |
"params": null,
|
|
|
7 |
"apply_qk_norm": true,
|
8 |
"attn_activation": null,
|
9 |
"attn_name": "auto",
|
|
|
12 |
"dim": 2048,
|
13 |
"ffn_type": "swiglu_torch",
|
14 |
"model": "open_lm_1b_swiglutorch",
|
15 |
+
"norm_type": "gain_only_lp_layer_norm",
|
16 |
"moe_capacity_factor": 1.25,
|
17 |
"moe_expert_model_parallelism": false,
|
18 |
"moe_freq": 0,
|
|
|
28 |
"qk_norm": true,
|
29 |
"seq_len": 2048,
|
30 |
"vocab_size": 50432,
|
31 |
+
"weight_tying": false,
|
|
|
32 |
"torch_dtype": "float32",
|
33 |
"transformers_version": "4.40.2"
|
34 |
}
|